How to Convert M4B to MP3 (Whole File or Chapter by Chapter)

What is an M4B file?

M4B is an audiobook container format developed by Apple. Under the hood it is just an MPEG-4 file (same as M4A) with a couple of extras: chapters baked into the file and a bookmark that saves your place. The audio itself is typically AAC, which is actually a better codec than MP3 at equivalent bitrates, so converting to MP3 does involve a small quality trade-off.

That said, MP3 is universally supported. Older car stereos, basic MP3 players, and a lot of third-party apps do not handle M4B at all. Converting makes sense if you want to listen on a device that will not touch Apple's format.

Add m4b-file-in-finder.webp next to this page's index.html

The easy way: convert in your browser

If you do not want to install anything or write a line of terminal commands, M4BConvert handles all of this in your browser. It runs ffmpeg locally via WebAssembly, so your file never leaves your machine, and it supports the full chapter-splitting workflow out of the box.

If you're staying for the terminal walkthrough, read on.

Install ffmpeg

Everything in this guide runs through ffmpeg (and its companion tool ffprobe), so install that first. None of the commands below will work without it. It takes about two minutes on any platform.

macOS

brew --version

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

brew install ffmpeg

Windows

Windows does not include ffmpeg, so you have to add it yourself. If you're on Windows 10 (2018 update or newer) or Windows 11, the fastest way is winget. If that does not work for you, use the manual method below. It works on every version of Windows.

Option 1: winget (fastest)

winget install ffmpeg

Let it finish, then close and reopen your terminal window and skip down to "Verify it worked" below. If you get "winget is not recognized," your Windows install is too old for it. Use Option 2 instead.

Option 2: manual install (works on any Windows version)

Linux

It's in your distro's package manager. Pick the matching command:

sudo apt install ffmpeg

sudo dnf install ffmpeg

sudo pacman -S ffmpeg

Verify it worked

Whichever method you used, confirm it from a fresh terminal window:

ffmpeg -version

You should see a version string and a long line of build flags, not "command not found" or "not recognized."

Add ffmpeg-version-check.webp next to this page's index.html

Check whether your file has chapters

Before you convert anything, find out whether your M4B actually has chapter markers, since that's what decides which option you'll use in the next section.

First, get the file's exact path instead of typing the filename

The single most common way this step "fails" has nothing to do with chapters. Typing just a filename, like audiobook.m4b, only works if your terminal happens to already be sitting in the exact same folder as the file, and it usually isn't, whether the file is in Downloads, on a USB drive, or somewhere custom. The fix that works no matter where the file lives is to copy the full path instead of typing any of it.

Windows

macOS

Linux

Add windows-copy-as-path.webp next to this page's index.html

Now run the chapter check

Paste the path you just copied in place of the filename below, keeping the quotes (Windows already added them; on macOS/Linux, add your own if the path has spaces in it):

ffprobe -v quiet -print_format json -show_chapters "PASTE_YOUR_COPIED_PATH_HERE"

For example, a real Windows path would look like:

ffprobe -v quiet -print_format json -show_chapters "D:\Downloads\wonderful_wizard_oz_version_4_1501.m4b"

This prints JSON with a chapters array. Each entry has a title, a start_time, and an end_time in seconds. That's everything needed to slice the file later.

Add ffprobe-chapter-json-output.webp next to this page's index.html

Convert M4B to MP3

Two options, depending on what the previous step told you.

Same rule as the chapter check: use the path you copied earlier instead of typing a filename, especially if the file isn't in your terminal's current folder.

Option A: single MP3 file

Use this if your file has no chapters, or if you genuinely just want one file. Paste your copied path in place of PASTE_YOUR_COPIED_PATH_HERE:

ffmpeg -i "PASTE_YOUR_COPIED_PATH_HERE" -q:a 0 output.mp3

For example:

ffmpeg -i "D:\Downloads\wonderful_wizard_oz_version_4_1501.m4b" -q:a 0 output.mp3

Here's what the flags mean:

• -i "...": the input file, your copied path in quotes
• -q:a 0: VBR quality setting, where 0 is the highest quality. Use -q:a 2 for a smaller file with barely noticeable difference, or -q:a 5 if size matters more than fidelity.
• output.mp3: ffmpeg infers the MP3 format from the file extension. This one's fine to leave as a plain filename, it'll save into whatever folder your terminal is currently in.

This puts the entire audiobook into one MP3 file. If it's 12 hours long, you get one 12-hour MP3 with no chapter markers your player can jump between. If that's a problem, use Option B instead.

Add ffmpeg-convert-command-output.webp next to this page's index.html

Option B: one MP3 per chapter

ffmpeg has no built-in "split by chapters" flag, so the script below loops through the chapter list from ffprobe and calls ffmpeg once per chapter to cut and encode it as its own MP3. It uses Python 3, which is built into macOS and most Linux distros, and available via the Microsoft Store on Windows.

Here's the script. Don't worry if you've never written or run a Python file before, the walkthrough right after it covers saving and running with nothing more than what's already on your computer.

#!/usr/bin/env python3
# m4b-to-chapters.py
# Splits an M4B audiobook into individual MP3s, one per chapter.
# Requires: ffmpeg + ffprobe on your PATH
#
# Usage:
#   python3 m4b-to-chapters.py audiobook.m4b
#   python3 m4b-to-chapters.py audiobook.m4b --quality 2

import argparse, json, os, re, subprocess, sys
from pathlib import Path

def probe_chapters(path):
    result = subprocess.run(
        ["ffprobe", "-v", "quiet", "-print_format", "json",
         "-show_chapters", str(path)],
        capture_output=True, text=True
    )
    return json.loads(result.stdout).get("chapters", [])

def safe_name(title):
    return re.sub(r"[^\w\s-]", "", title).strip().replace(" ", "_")[:80]

def convert(m4b_path, quality=0):
    src      = Path(m4b_path)
    chapters = probe_chapters(src)

    if not chapters:
        print("No chapter markers found. Converting as a single file.")
        out = src.with_suffix(".mp3")
        subprocess.run(["ffmpeg", "-y", "-i", str(src),
                        "-q:a", str(quality), str(out)], check=True)
        return

    out_dir = src.parent / (src.stem + "_chapters")
    out_dir.mkdir(exist_ok=True)
    print(f"Found {len(chapters)} chapters → {out_dir}/\n")

    for ch in chapters:
        i     = int(ch["id"]) + 1
        title = ch["tags"].get("title", f"Chapter {i}")
        start = ch["start_time"]
        end   = ch["end_time"]
        fname = f"{i:02d}_{safe_name(title)}.mp3"
        out   = out_dir / fname

        subprocess.run([
            "ffmpeg", "-y",
            "-ss", start, "-to", end,
            "-i", str(src),
            "-q:a", str(quality),
            "-map_metadata", "-1",
            "-metadata", f"title={title}",
            "-metadata", f"track={i}",
            str(out)
        ], capture_output=True, check=True)
        print(f"  converted: {fname}")

    print(f"\nDone. {len(chapters)} MP3s saved to {out_dir}/")

if __name__ == "__main__":
    p = argparse.ArgumentParser(description="Split M4B into chapter MP3s")
    p.add_argument("input", help="Path to .m4b file")
    p.add_argument("--quality", type=int, default=0,
                   help="VBR quality: 0 (best) to 9 (smallest). Default: 0")
    args = p.parse_args()
    convert(args.input, args.quality)

Save the script as a .py file

No need for VS Code or any code editor here. Any plain text editor will do, including whatever's already built into your computer. The only thing to watch for is two small traps: Notepad on Windows quietly tacks on a .txt extension unless you stop it, and TextEdit on Mac saves rich text by default, which will corrupt the script without you noticing. Both are a non issue once you know they're there.

Windows (Notepad)

macOS (TextEdit)

Linux (any text editor)

A GUI editor like GNOME Text Editor, gedit, or Kate saves plain text by default, so just open one, paste the script, and save it as m4b-to-chapters.py. No traps to worry about here. If you'd rather stay in the terminal:

nano m4b-to-chapters.py

Paste the script in (usually right-click and Paste, or Ctrl+Shift+V), then press Ctrl+O to save and Ctrl+X to exit.

Run it

python3 "PASTE_SCRIPT_PATH_HERE" "PASTE_AUDIOBOOK_PATH_HERE"

python3 "C:\Users\ASUS\Desktop\m4b-to-chapters.py" "D:\Downloads\wonderful_wizard_oz_version_4_1501.m4b"

Add chapterization-script-running.webp next to this page's index.html

Add chapters-folder-output.webp next to this page's index.html

Load the folder onto your device and any MP3 player will see them as separate tracks in order. Most car stereos and dedicated MP3 players handle a folder of sequentially named files exactly like chapters.

Troubleshooting

ffprobe returns {} with no chapters, but I know the file has them

This is almost always a file-not-found error hiding behind -v quiet, not actually missing chapter data. ffprobe suppresses the "no such file or directory" message along with everything else when that flag is set, so it looks identical to a real empty result. Confirm the file is actually where you think it is (dir *.m4b on Windows, ls *.m4b on macOS/Linux), or just use the full path, e.g. "D:\Downloads\audiobook.m4b" if your Downloads folder lives on a different drive. Dropping -v quiet for one test run will show you the real error instead of a silent {}.

The output MP3 has no chapters / just plays as one track

That's expected from Option A (single file). Use Option B above, or M4BConvert, to get individual MP3s per chapter.

The script says "No chapter markers found"

Some M4B files genuinely have no chapter table, especially ones converted from audiobook CDs or ripped from DRM-free downloads without chapter data. Run ffprobe -v quiet -print_format json -show_chapters yourfile.m4b to confirm. If the chapters array is empty, the file has no chapter data and the script will convert the whole thing to a single MP3 automatically.

Poor audio quality in the output

Make sure you're using -q:a 0 or at most -q:a 2. Higher numbers mean lower quality. Also, the source M4B is usually encoded at 64 kbps (common for speech audiobooks), so there's a ceiling to how good the output can sound regardless of the encoder settings. Converting at 64 kbps AAC input, you won't get CD-quality MP3 out; you'll get a clean re-encode of whatever was there to begin with.

Conversion is very slow on a large file

Option A should be fast (often faster than real-time on modern hardware). Option B calls ffmpeg once per chapter, which adds some overhead, but should still be quick. If it's crawling, check that ffmpeg isn't also transcoding video; use ffprobe yourfile.m4b to confirm it's audio-only.

Windows says to install Python from the Microsoft Store

That message comes from a Windows placeholder called an App execution alias, not from Python itself, and it shows up the moment you type python3 if Python isn't registered under that exact name. If you don't have Python yet, installing from the Store works fine. If you installed it from python.org instead, that installer registers the command as py, so use py m4b-to-chapters.py "your path here" instead of python3. You can also turn the placeholder off entirely at Settings, then Apps, then Advanced app settings, then App execution aliases.

Python says "ModuleNotFoundError" or the script won't run on Windows

The script only uses standard library modules, so there's nothing to pip install. Make sure you have Python 3.6 or later, and try py in place of python3 if that's what your install responds to (see the entry above). If you're on macOS or Linux, python3 should already work out of the box.

"ffmpeg is not recognized" even after installing on Windows

This almost always means PATH was updated but your terminal window is still the old one. Close every open Command Prompt / PowerShell window and open a fresh one. PATH changes only apply to new windows. If it's still not found after that, double check the exact folder path you added in the install step actually contains ffmpeg.exe.